vouh-chronos-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| WMC_PYTHON | No | Path to the Python interpreter used to run `py/vision_worker.py`. | auto-detected (`python`, `py`, `python3`) |
| FFMPEG_PATH | No | Custom path to the `ffmpeg` binary. | @ffmpeg-installer/ffmpeg |
| WMC_WORKDIR | No | Root directory where captures are stored (`<WORKDIR>/<capture_id>/`). | ./wmc-data |
| FFPROBE_PATH | No | Custom path to the `ffprobe` binary. | @ffprobe-installer/ffprobe |
| WMC_DELIVERY | No | Default delivery mode for `get_frames`: `images`, `sheets`, `zip`, or `video`. | images |
| WMC_ALLOWED_HOSTS | No | Comma-separated list of allowed hosts (e.g. `localhost,127.0.0.1,my-site.test`). | localhost,127.0.0.1 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| capture_siteB | Ouvre l'URL dans Chromium (Playwright), joue un scénario d'interaction, enregistre la vidéo .webm et la timeline instrumentée. |
| extract_framesB | Découpe la vidéo d'une capture en frames JPEG numérotées depuis 0 avec leur index horodaté exact (index.json, index.md). |
| get_framesB | Renvoie les frames extraites par lots paginés (images JPEG, planches contact, archive ZIP ou chemin vidéo) avec leur index horodaté. |
| analyze_motionB | Calcule les durées, courbes d'easing (Bézier / ressort), parallax, lissage de scroll, cascades (stagger) et flux optique, puis écrit motion-spec.md. |
| build_reproduction_packB | Prépare le dossier reproduction_pack/ contenant layout.json (blocs neutres), tokens.json, motion-tokens.json, motion-spec.md, frames/ ou sheets/, et README.md. |
| compare_capturesB | Compare deux captures (alignement DTW sur l'énergie de mouvement, SSIM sur les frames alignées, écart de durée à ±1 frame et RMSE des courbes). |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| reproduce_interface | Guide l'agent pour reconstruire la structure géométrique et le mouvement d'une capture de référence avec un contenu strictement neutre (§11.2). |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
The tools mostly target distinct workflow stages: capture, frame extraction/retrieval, motion analysis, reproduction packaging, and comparison. The only potential confusion is between `get_frames` and `extract_frames`, since one extracts frames and the other retrieves them, but the descriptions distinguish extraction from paginated retrieval.
All tool names follow a consistent snake_case verb_noun pattern: capture_site, get_frames, extract_frames, analyze_motion, build_reproduction_pack, compare_captures. The convention is predictable and readable throughout.
Six tools is well-scoped for a capture-to-analysis pipeline. Each tool corresponds to a clear stage in the workflow, with no obviously redundant or missing bulk.
The toolset covers the core lifecycle: capture a site, extract and retrieve frames, analyze motion, build a reproduction pack, and compare captures. Minor gaps exist around managing or listing prior captures and deleting artifacts, but the primary workflow is complete.